Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    diver

    np_max-act · gemini-2.0-flash

    discussions related to software versioning and compatibility issues.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects non‐natural “control” tokens and markup elements (e.g. header or metadata tags) rather than ordinary words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ۵۰
    -0.07
    427
    -0.06
    ремя
    -0.06
    xCB
    -0.06
    EHICLE
    -0.06
    iado
    -0.06
     defendants
    -0.06
     overridden
    -0.06
     мінім
    -0.06
    owane
    -0.06
    POSITIVE LOGITS
     joking
    0.07
     ek
    0.07
    bios
    0.06
    atial
    0.06
    >↵↵↵↵↵
    0.06
     olmam
    0.06
     Vertical
    0.06
    	throw
    0.06
     harmon
    0.06
     vh
    0.06
    Act Density 0.637%

    No Known Activations